Changes in the use of the subjunctive

Here’s a grammar question for you. Which form of the verb be would you use in the first sentence below, and which form of the verb take in the second?

I wish that Kirsty … here to celebrate with us.

It’s essential that Sam … his phone with him.

Prescriptive grammars, which tell you how you should and shouldn’t use language, will say that you must use were in the first sentence and take in the second.

I wish that Kirsty were here to celebrate with us.

It’s essential that Sam take his phone with him.

oup_56043_croppedThe forms were and take in the second pair of sentences are often called subjunctive verb forms which indicate that a particular situation is unreal or not the case (Kirsty is not at the party; Sam hasn’t (yet) taken his phone with him), but is nevertheless wished-for. The second example above illustrates the use of the so-called mandative subjunctive to indicate the importance or necessity of something happening. Subjunctive verb forms will be familiar to you if you speak one or more of the romance languages, such as Spanish, Italian, or Portuguese.

But are were and take the only correct forms in the sentences above? More descriptively oriented grammar books will tell you that you can also use was and takes, or should take, at least in British English:

I wish that Kirsty was here to celebrate with us.

It’s essential that Sam takes his phone with him.

It’s essential that Sam should take his phone with him.

The first sentence uses the regular third person past tense form of the verb be, and the second uses the regular third person present tense form of take. In the third sentence we have an example of mandative should, which offers an alternative way of expressing necessity. These sentences would sound perfectly normal to many (especially younger) British speakers, but they would sound ungrammatical, or at least unusual, to many American ears.

Writing in the early part of the twentieth century, the Fowler brothers famously claimed in their book The King’s English that the mandative subjunctive should be avoided because it can be ‘dangerous’ (!) and is often ‘unpleasantly formal’. In any case, they argued, the subjunctive is unnecessary and about to disappear from the English language. It turned out that the Fowlers were wrong, and that not only did the subjunctive survive, it had a revival in British English during the second part of the twentieth century. For some, this caused anxiety, as this passage from Catherine Nesbitt from the early 1960s shows:

Today I would like to draw attention to something far more serious, the unexpected revival of the Subjunctive Mood, which seems to have begun in this country less than ten years ago and is now spreading so rapidly that, if left unchecked, it will do real damage to the structure of the language, a far more harmful thing than any craze for the latest fashionable word.

The revival of the subjunctive is quite surprising because to many modern ears it does sound rather quaint and formal. So what could be the reasons for it? Linguists have speculated that it may have happened under the influence of American English, in which the subjunctive was always more frequent. American English has been influencing British English ever since films, music and television programmes made their way across the ocean. However, some recent research suggests that the increased use of the subjunctive has now stalled. Who knows, in the longer term maybe the Fowlers may still be proved right about the subjunctive disappearing from the English language.

Bas Aarts is Professor of English Linguistics at UCL and the author of Oxford Modern English Grammar. He is a founder of the Englicious website which contains free English language teaching resources and writes the blog Grammarianism for teachers of English.

Modal verbs: how their use has declined in recent times

mustAs is well-known, English has a number of core modal verbs, including can/could, will/would, shall/should, may/might, and must. These make it possible for us to express meanings such as ‘possibility’, ‘necessity’, ‘obligation’, ‘permission’, and the like. In an earlier blog post I discussed the declining use of the modal verb must over recent decades. A very likely reason for this is that speakers have become less willing to appear to be telling others what to do, even in situations when they are entitled to do so.

What about the other modal verbs in English? The table below shows how their use has declined across the board in written and spoken British English between the 1960s and the 1990s:*

Changes in modal usage, calculated as a percentage change per million words of text. Bold figures are statistically significant.

Notice first of all that the rate of change is different for the various modals in written and spoken English. For example, the use of can has increased more in spoken than in written English. This may be due to speakers using this verb more often to ask for permission than may, which has declined in use.

Let’s now take a closer look at shall, which is much more common in British English than in American English. Like must, it has also dropped dramatically in use. When used with first person subjects shall alternates with will:

I shall look into this.

I will look into this.

Is there a difference between these two ways of talking about a future event? From a stylistic point of view there is: the first of these alternatives sounds more formal, and is more likely to be used by older speakers. However, there is really no difference in meaning: both sentences are used to speak about a situation in the future.

Not everyone would agree with what I just wrote. Some writers would say that with first person subjects (I/we) will should be used to convey the idea of ‘wanting’, whereas shall should be used to talk about a neutral future. With other persons (he, you, they, etc.) shall should be used for promises/guarantees (Cinderella shall go to the ball), in ‘regulatory language’ (Students shall not enter the premises after midnight), or to express something that the speaker wants (You came for the action, and action you shall have). The journalist Simon Heffer explains the difference with a well-known anecdote in his book Strictly English (2012):

The Victorian schoolmaster had a way of impressing this distinction upon his charges, with the story of the boy who drowned: for he had cried out ‘I will drown, and no-one shall save me’.

Heffer regrets the fact that speakers no longer know how to use shall and will in the right way. Maybe it’s true that a useful distinction in English is lost. However, another way of looking at this development is to recognize that the language has changed in such a way that the difference between shall and will (other than in questions) is slowly disappearing because speakers no longer sense a meaning difference between these verbs. This would mean that speakers need only one verb, and it looks like will has won out. The increased use of will shown in the table above supports this hypothesis, but the increase is not as high as you would expect, so perhaps the table only offers a partial explanation for the trends in usage. What’s interesting is that the decline of shall is slightly higher in spoken English than in written English. This may well mean that the decline will spread in the future because changes in spoken language often make their way into written language.

Bas Aarts is Professor of English Linguistics at UCL and the author of Oxford Modern English Grammar. He is a founder of the Englicious website which contains free English language teaching resources and writes the blog Grammarianism for teachers of English.

* The data are from Bas Aarts, Jill Bowie and Sean Wallis (2015) Profiling the English verb phrase over time: modal patterns. In: Irma Taavitsainen, Merja Kytö, Claudia Claridge and Jeremy Smith (eds.) Developments in English: expanding electronic evidence. (Cambridge: Cambridge University Press) and from Geoffrey Leech, Marianne Hundt, Christian Mair, and Nicholas Smith (2015) Change in contemporary English: a grammatical study (Cambridge: Cambridge University Press).


Languages change all the time

If you have studied the English language at school, college or university you will know that it has changed since its beginnings to the present time. The following are the opening lines of Beowulf, written in Old English between 700 and 1000:

Hwæt! We Gardena in geardagum
þeodcyninga, þrym gefrunon
hu ða æþelingas ellen fremedon

Unless you have taken a course in Old English, it is quite difficult, if not impossible, to understand this passage. English gets easier to understand over time, so that Chaucer’s Middle English is not as hard to read as Old English, and Shakespeare’s Early Modern English is not all that far removed from Modern English.

The Swiss linguist Ferdinand De Saussure famously made a distinction between studying a language diachronically, i.e. over time, and synchronically, i.e. at a particular time in its history. The distinction seems to be a useful one when we compare Beowulf with any piece of modern writing. However, in recent times linguists have questioned De Saussure’s distinction, and have pointed out that languages change all the time, even within the life span of individuals, and we can therefore speak of changes in progress, sometimes called ‘current change’. Nowadays, with the advent of large collections of language data (called corpora) tracking such changes has become easier.

Languages can change in many different ways.  The kinds of changes we can observe include sound changes, changes in the meaning or frequency of use of words, and changes in grammar. Let’s have a look at some examples.

Sound changes played a very prominent role in English, especially between 1350 and 1600 when the Great Vowel Shift took place. However, even in the present time changes in pronunciation happen. This is very obvious when you listen to announcers in old cinema newsreels: they sound distinctly different from present-day newscasters. Even the Queen’s pronunciation has changed over recent years, as recent research from a German University has shown.

In a post on my blog Grammarianism I discuss how the use of modal verbs has changed over the past few decades. Especially notable is the decline of must, which we can use to tell others to do something (“You must arrive by 6 p.m. at the latest”). Most likely as a result of our society becoming less hierarchical, speakers of English now tend to use different ways of making requests of others, without sounding too authoritarian. For example, they might use have to (“You have to arrive by 6 p.m. at the latest”) or need to (“You need to arrive by 6 p.m. at the latest”) instead.

As another example of changes in the use of particular words Catherine Soanes writes about the decline of the word whilst on the Oxford Dictionaries blog. She suggests that the reason for this decline might be that it sounds somewhat old-fashioned and pretentious, at least in the US. As for the UK, she observes:

In British English, whilst’ incurs less opprobrium, but guides and dictionaries usually advise that ‘while’ is preferable, given that it’s the most common form and may sound more up to date.

My own observations suggest that whilst may be making a comeback. I heard it recently in an announcement on a train platform: “Please take care whilst it is raining”. More tellingly, I hear it all the time in the speech of my 9-year old daughter and her friends. Of course, this is only anecdotal, and not solid evidence of a change, but if younger generations start re-using words, they may well be coming back into the language. Think also of the word cool which wasn’t cool for a long time, but it is now again being used by young people. Returning to whilst, interesting data on its use in the US are available in the following table from the Corpus of Contemporary American English:


COCA contains 520 million words, and the data are sourced from a large number of different text categories, including spoken material (shown at the top of the table). In the right-hand portion of the table we see a slight, but steady, increase in the use of whilst per million words between 1995 and 2015.

What about changes in grammar, i.e. the syntactic patterns that we find in language? Another observation I made in the speech of my youngest daughter (she’s a great source of data!) is the emergence of what I will call the isn’t it that-construction. Here are some examples:

Dad, isn’t it that you promised to make pancakes?

Isn’t it that Sarah’s party is this afternoon?

This construction seems to constitute a new way of asking questions: rather than asking Dad, didn’t you promise to make pancakes?  the new construction seems to be a clipped version of the longer Isn’t it the case that you promised to make pancakes? It’s not clear how the new construction came about, or whether it is more widespread than just in the speech of my daughter and her friends. At present the most we can say is that changes like this are trends, and more research is needed to find out if they will ever fully become part of the English language.

Bas Aarts is Professor of English Linguistics at UCL and the author of the Oxford Modern English Grammar. He is also a founder of the free Englicious website for teachers of English.